Intensity estimation on geometric networks with penalized splines

DOI10.1214/21-AOAS1522MaRDI QIDQ2154182zbMATH OpenOpenAlexFDO

Authors Marc Schneble, Göran Kauermann

Publication date 14 July 2022

Published in The Annals of Applied Statistics (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/2002.10270

zbMATH Keywords

generalized additive modelspenalized splinesgeometric networksintensity estimation of stochastic point processesPoisson regression with offset

Mathematics Subject Classification ID

Computational methods for problems pertaining to statistics (62-08) Density estimation (62G07) Nonparametric regression and quantile regression (62G08) Point processes (e.g., Poisson, Cox, Hawkes processes) (60G55) Generalized linear models (logistic models) (62J12)

Abstract: In the past decades, the growing amount of network data has lead to many novel statistical models. In this paper we consider so called geometric networks. Typical examples are road networks or other infrastructure networks. But also the neurons or the blood vessels in a human body can be interpreted as a geometric network embedded in a three-dimensional space. In all these applications a network specific metric rather than the Euclidean metric is usually used, which makes the analyses on network data challenging. We consider network based point processes and our task is to estimate the intensity (or density) of the process which allows to detect high- and low- intensity regions of the underlying stochastic processes. Available routines that tackle this problem are commonly based on kernel smoothing methods. However, kernel based estimation in general exhibits some drawbacks such as suffering from boundary effects and the locality of the smoother. In an Euclidean space, the disadvantages of kernel methods can be overcome by using penalized spline smoothing. We here extend penalized spline smoothing towards smooth intensity estimation on geometric networks and apply the approach to both, simulated and real world data. The results show that penalized spline based intensity estimation is numerically efficient and outperforms kernel based methods. Furthermore, our approach easily allows to incorporate covariates, which allows to respect the network geometry in a regression model framework.

Recommendations

• On Kernel-Based Intensity Estimation of Spatial Point Patterns on Linear Networks
• Spline interpolation on networks
• On the asymptotics of penalized splines
• Cubic \(L_1\) splines on triangulated irregular networks
• Spline density estimation and inference with model-based penalties
• Asymptotic theory of penalized splines
• Adaptive estimation of nonparametric geometric graphs
• On the asymptotics of penalized spline smoothing

Cites work

• scientific article; zbMATH DE number 6734253 (Why is no real title available?)
• A Kernel Method for Smoothing Point Process Data
• A generalized Fellner-Schall method for smoothing parameter optimization with application to Tweedie location, scale and shape models
• A survey of statistical network models
• All-pairs shortest paths for unweighted undirected graphs in \(o(mn)\) time
• Data-driven selection of the spline dimension in penalized spline regression
• Density estimation and comparison with a penalized mixture approach
• Fast Kernel Smoothing of Point Patterns on a Large Network using Two‐dimensional Convolution
• Flexible smoothing with B-splines and penalties. With comments and a rejoinder by the authors
• Fused Density Estimation: Theory and Methods
• Generalized Linear Array Models with Applications to Multidimensional Smoothing
• Geometrically corrected second order analysis of events on a linear network, with applications to ecology and criminology
• Kernel density estimation on a linear network
• Multivariate density estimation. Theory, practice, and visualization
• On Kernel-Based Intensity Estimation of Spatial Point Patterns on Linear Networks
• On a small sample adjustment for the profile score function in semiparametric smoothing models
• On calculating with B-splines
• On the Voronoi estimator for the intensity of an inhomogeneous planar Poisson process
• Point processes on directed linear networks
• Regression. Models, methods and applications
• Resample-smoothing of Voronoi intensity estimators
• Semiparametric Regression
• Semiparametric regression during 2003--2007
• Statistical analysis of network data with R
• Stochastic actor‐oriented models for network change
• The structure of extended real-valued metric spaces
• Twenty years of P-splines (invited article)
• Validation and comparison of geostatistical and spline models for spatial stream networks

Cited in

(3)

• Geometrically corrected second order analysis of events on a linear network, with applications to ecology and criminology
• Kernel density estimation on a linear network
• Density estimation on a network

Describes a project that uses

Uses Software

• spatstat
• R

This page was built for publication: Intensity estimation on geometric networks with penalized splines

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2154182)